madlad400-finetuned-lag-swh

This model is a fine-tuned version of facebook/nllb-200-distilled-1.3B for translation from Rangi to Swahili.

Model details

• Developed by: SIL Global
• Finetuned from model: facebook/nllb-200-distilled-1.3B
• Model type: Translation
• Source language: Rangi (lag)
• Target language: Swahili (swh)
• License: closed/private

Datasets

The model was trained on a parallel corpus of plain text files:

Rangi:

• Rangi New Testament
• License: All rights reserved, Wycliffe Bible Translators. Used with permission.

Swahili:

• Swahili back-translation of Rangi New Testament
• License: All rights reserved, Wycliffe Bible Translators. Used with permission.

Framework versions

• PEFT 0.12.0
• Transformers 4.44.2
• Pytorch 2.4.1+cu124
• Datasets 2.21.0
• Tokenizers 0.19.1

Usage

You can use this model with the transformers library like this:

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("sil-ai/madlad400-finetuned-lag-swh")
model = AutoModelForSeq2SeqLM.from_pretrained("sil-ai/madlad400-finetuned-lag-swh")

inputs = tokenizer("Your input text here", return_tensors="pt")
outputs = model.generate(**inputs)
print(tokenizer.decode(outputs[0]))